Biochemistry 1988, 27, 5763-5771
5763
Processivity of Proteolytically Modified Forms of T7 RNA Polymerase? Daniel K. Muller,* Craig T. Martin,# and Joseph E. Coleman* Department of Molecular Biophysics and Biochemistry, Yale University, New Haven, Connecticut 06510 Received December 15, 1987; Revised Manuscript Received March 17, I988
ABSTRACT: Two proteolytically modified forms of T7 R N A polymerase have been characterized with respect to transcription initiation and processivity. One species, denoted 80K-20K, is singly cleaved within the region of the polypeptide chain between amino acids 172 and 180. The second species, denoted 80K, is generated by extensive proteolysis of the N-terminal 20K domain by trypsin. The 80K-20K form is fully active in initiation and escape from abortive cycling. It is deficient only in processivity on long D N A templates. Likewise, the 80K species shows initiation kinetics and abortive product synthesis similar to those of the native enzyme. This latter species, however, is unable to escape abortive cycling and shows no synthesis of transcripts longer than about eight bases. Studies of R N A and DNA binding to the three different forms of the enzyme by gel retention assays reveal that the native (98K), the 80K-20K, and the 80K species all form specific complexes with promoter-containing DNA. In addition, the native enzyme binds nonspecifically to double-stranded DNA, while the 80K-20K and 80K enzymes do not. The native enzyme also binds RNA. This R N A binding is reduced in the 80K-20K enzyme and is absent in the 80K species. We suggest a model for T7 R N A polymerase wherein the 20K N-terminal domain of the protein or a shared region between the N- and the C-terminal domains of the protein forms a nonspecific polynucleotide binding site. Binding of the nascent mRNA to such a site may be involved in the observed transition of the native enzyme from the newly initiated complex, susceptible to abortive cycling, to a more stable highly processive ternary complex.
R N A polymerase from the bacteriophage T7 is a small monomeric enzyme which is highly specific for transcription of a set of small, well-conserved promoters within the T7 genome (Chamberlin et al., 1970; Chamberlin & Ring, 1973; Niles et al., 1974; Oakley & Coleman, 1977; Dunn & Studier, 1981). In contrast to the large multisubunit bacterial or eukaryotic RNA polymerases, the T7-encoded enzyme is much less complex both structurally and functionally and requires no zinc or external cofactors for transcription (King et al., 1986). With the cloning and overexpression of T7 gene 1 (the polymerase gene), large amounts of the enzyme have become available for protein chemical studies. These facts combine to make the T7 system an excellent choice in which to study the mechanisms of transcription. Proteolysis of cloned T7 RNA polymerase has been observed during purification and results in the nicking of the enzyme to produce an 80K- and a 20K-size fragment (Davanloo et al., 1984; Tabor & Richardson, 1985). In a similar fashion, the RNA polymerase from the related bacteriophage T3, which is over 80% homologous to the T7 enzyme, has also been observed to undergo proteolysis (Kupper, 1974). For both enzymes it has been observed that this nicking results in a loss of approximately 80% of the polymerase activity as measured in the standard assay (Oakley et al., 1975). We have recently developed a kinetic assay for initiation by T7 RNA polymerase which allows the direct determination of Michaelis-Menten kinetic parameters from transcription of a short oligonucleotide promoter template (Martin & Coleman, 1987). Initiation kinetics for the native enzyme fit well to the simple steady-state Michaelis-Menten equation: ‘This work was supported by NIH Grant GM21919-13. tD.K.M. was supported by National Institutes of Health Training Grant GM07223. C.T.M. is the recipient of National Institutes of Health Postdoctoral Fellowship GM10902.
*
0006-2960/88/0427-5763$01 SO10
enzyme
-
+ DNA K.T enzyme-DNA enzyme
+ kcat
DNA
+ message
(1)
The rate constant, k,,, reflects the slowest step or steps in the initiation process. In this assay the production of a well-defined five-base message avoids contributions to the observed incorporation rate from the elongation phase of transcription. Since the five-base message is the predominant product, complications in the measurement of transcription initiation due to premature termination are avoided. In the current study, we have exploited the single-site proteolysis of T7 RNA polymerase by whole Escherichia coli cells (Dunn and Studier, personal communication) to produce a homogeneous preparation of uniquely cleaved “80K-20K species analogous to that recently studied by Ikeda and Richardson (1987a,b). The latter study concluded that the reduced activity of the singly cleaved enzyme was due to both a lower initiation rate and premature chain termination. We now show that single-site cleavage between the 80K and 20K polypeptides does not affect specific promoter recognition or initiation of transcription; however, the “processive complex” shows decreased processivity relative to the native form. In addition, we have exploited proteolysis by trypsin to generate a new form of the enzyme, denoted “80K”, in which the 20K N-terminus of the protein is extensively digested. This proteolyzed form of the enzyme has an only slightly reduced initiation rate, suggesting that the catalytic determinants of transcription in T7 RNA polymerase reside solely in the larger C-terminal polypeptide. We have recently characterized abortive cycling by native T7 RNA polymerase (Martin et al., 1988). Under normal transcription conditions the enzyme produces many small transcripts (less than eight bases long) for each full-length transcript made. We observe here that all three species of the enzyme synthesize the abortive products but only the native and 80K-20K enzymes produce full-length transcripts. In the 0 1988 American Chemical Society
5764 B I O C H E M I S T R Y 80K species, the transition from an initiated to a fully processive ternary complex is inhibited. Comparison of these two proteolytically modified forms of T7 RNA polymerase with the native enzyme provides valuable insight into the domain structure and mechanism of T7 RNA polymerase. MATERIALS AND METHODS Purification. T7 RNA polymerase was purified from E . coli strain BL21 containing plasmid pAR1219 (kindly supplied by William Studier and John Dunn, Brookhaven National Laboratories), according to W. Studier as described by King et al. (1986). A molar extinction coefficient of e280 = 1.4 X lo5 M-' cm-I was used to determine enzyme concentrations (King et al., 1986). Both the preparation of oligonucleotide and the kinetic assays of transcription on the oligonucleotide template were performed as previously described (Martin & Coleman, 1987). Concentration of oligonucleotide and plasmid DNA was determined by the absorbance at 260 nm by assuming an optical density of 1.OO represents 50 pg/mL double-stranded DNA. Double-stranded syntheticoligonucleotides were prepared by heating the single strands to 90 OC and allowing the solution to slowly cool to room temperature. Proteolysis by Whole Cells. T7 RNA polymerase was digested according to Dunn and Studier (personal communication) to produce the singly-cleaved 80K-20K enzyme form. Briefly, E. coli strain HMS174 was grown overnight on agar plates with Luria broth. The cells were then suspended in 50% glycerol to form a stock solution. The cleavage reaction was carried out in 20 mM potassium phosphate, 1 mM EDTA,' and 150 mM NaCl with HMS174 cells added to a final ODm, of 1.5. The mixture was then incubated at 37 OC for approximately 6 hr. The reaction was stopped by centrifugation and filter sterilization of the supernatant. The extent of the cleavage was monitored by SDS-PAGE. We have found this procedure to produce uniform samples of the singly cleaved enzyme. Proteolysis by Trypsin. Digestion of the enzyme with TPCK-trypsin (Worthington) was carried out in 20 mM potassium phosphate, l mM EDTA, and 100 mM NaCl. To produce the 80K enzyme form, a 1:2000 (w/w) ratio of trypsin to protein was incubated at room temperature. Aliquots were removed at the indicated times and spotted into tubes containing a twofold excess of soybean trypsin inhibitor (Sigma) over trypsin. Analysis of RNA Transcripts. Transcripts produced from whole T7 DNA or from restriction fragments of T7 DNA were analyzed by electrophoresis on a composite gel system according to Chamberlin et al. (1979). The transcripts were made radioactive by including [cY-~*P]UTP (ICN) in the reaction mixture. The reaction was stopped by the addition of 5 X TAE buffer (1X TAE buffer is 40 mM Tris-HC1, 20 mM sodium acetate, and 2 mM EDTA, pH 7.5) containing 0.1% SDS, and an aliquot was loaded directly onto a 0.4 mm thick 0.5% agarose-2.75% polyacrylamide-0.08% bis(acry1amide) gel buffered in 1X TAE. Small transcripts produced from synthetic oligonucleotides or restriction digests of pIBI24, a T7 promoter containing plasmid (International Biotechnologies Inc., New Haven, CT), were analyzed by electrophoresis on a 0.2 mm thick, 20% acrylamide-7 M urea gel system in 0.5X TBE buffer (where 1X TBE buffer is 89 mM Tris, 89 mM boric acid, and 20 mM EDTA, pH 8.0). After a 20-min Abbreviations: EDTA, ethylenediaminetetraacetic acid; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis; Tris, tris(hydroxymethy1)aminomethane.
MULLER ET AL.
incubation in the standard assay buffer (Oakley et al., 1975) with 70 nM enzyme and 110 nM DNA, the samples were made 1 M in ammonium acetate, and 2 pg of tRNA was added. The transcripts were precipitated with 4 volumes of ethanol. After centrifugation the samples were dried and redissolved in 7 M urea (Bethesda Research Laboratories). Alternatively, the reaction was quenched by the direct addition of an equal volume of 90% formamide, 50 mM EDTA, and 0.01% bromphenol blue. Both methods produced the same results. In either case, the samples were heated to 90 O C , quick cooled, and then loaded directly onto a preelectrophoresed gel. All denaturing gels were electrophoresed at approximately 1200 V. These and subsequent gels were vacuum dried and autoradiographed. Gel Retention Assay To Monitor DNA Binding. Binding between enzyme and DNA was analyzed with a nondenaturing polyacrylamide gel system [for a review, see Hendrickson (1985)l. Electrophoresis of a mixture of protein and DNA results in the separation of free DNA from that bound to protein and hence provides a qualitative assay for DNA binding. The protein-DNA mixtures were incubated at 37 "C at concentrations of protein and DNA as indicated in Figure 3. All binding experiments were carried out in 10 mM potassium phosphate, pH 7.8, 1 mM EDTA, 20 mM NaC1, and 4% glycerol. The reaction mixtures were incubated for 10 min, loaded onto a preelectrophoresed 8% gel, and run at room temperature at 150 V. Both the gel and running buffer contained 1X TBE buffer. Purijication of RNA. A double-stranded synthetic 37 base pair oligonucleotide (see Results) was used as a template to produce large amounts of 20-base RNA transcript. The transcription reaction was carried out in 400 pL of the standard reaction buffer except that all four NTPs were made 2 mM instead of 0.4 mM. For the production of radioactive RNA, 40 pCi of [ c Y - ~ ~ P I Uwas T Padded. The reaction was carried out at 37 OC for 30 min. After the addition of glycerol and bromphenol blue dye, the reaction mixture was directly loaded onto a 20% nondenaturing polyacrylamide gel buffered in 0.5X TBE. The gel was autoradiographed to locate the 20-base RNA band, which was cut out and electroeluted according to Maniatis et al. (1982). The sample was made 1 M in ammonium acetate, and 4 volumes of 95% ethanol was added to precipitate the RNA. After centrifugation, the pellet was redissolved in 50 pL of 10 mM Tris-1 mM EDTA, pH 7.8, buffer. The concentration was measured by specific activity, if 32P-labeled nucleotides were used in the transcription reaction, or by the absorbance at 260 nm. The two measurements generally agreed within 10%. RESULTS Proteolysis of T7 RNA Polymerase. The specific cleavage of T7 RNA polymerase into 80K and 20K fragments during purification of the overproduced enzyme from E . coli strain HMSl74/pAR1219 has been previously reported (Davanloo et al., 1984). Native (98K) T7 RNA polymerase can be digested in a controlled manner into two polypeptide fragments with molecular weights of approximately 80K and 20K by incubation with E. coli HMS174 cells at 37 OC as shown in Figure 1A. To determine the exact position of the cut site, we analyzed the fully cleaved material by gas-phase peptide sequencing. Although sequencing is complicated by the presence of the original N-terminus of the protein, two sequences can be clearly identified and show that the bacterial protease cleaves after Lys-179, in agreement with the cleavage site determined by Dunn and Studier (unpublished observations). We have not observed any significant cleavage at other
D O M A I N S O F T7 R N A P O L Y M E R A S E
.
.--"-p.rrrcr-.
20K
20K
RNA polymerase. (A) Whole E. coli (strain HMS174) cells were added to purified T7 R N A polymerase. After 7-h incubation, the enzyme is separated into 80K and 20K polypeptides by SDS-PAGE ( I 3%) due to a single peptide bond cleavage after Lys-179. (B) Digestion of T7 R N A polymerase with trypsin. After 15 min of incubation with trypsin, the enzyme is separated into 80K and 20K polypeptides by SDS-PAGE ( 1 5%) due to peptide bond cleavage after Arg-173 and Lys-I80 (see text). After 1.5 h there is further digestion with the disappearance of the intact 20K polypeptide. The 80K species shows a N-terminal Ala ( 18 1 ) and has undergone some minor additional cleavage. This may be cleavage near the C-terminus (see text). Lane A contains a standard for 80K and 20K polypeptides, produced by long-term incubation with whole cells. FIGURE1: Time course of proteolysis of T7
sites within the protein when the purified enzyme is digested with whole E. coli strain HMS174. Tabor and Richardson (1985) have also reported cleavage of the enzyme during purification from E. coli strain HMS273, but after Lys-172. We have found that some cleaved enzyme is produced upon cell lysis and purification of the enzyme from E. coli BL21. In this case the cleavage occurs in this same region of the polypeptide chain but after Tyr-178 rather than Lys-179 as shown by gas-phase peptide sequencing. Similar proteolysis of T7 RNA polymerase can be achieved by digestion with trypsin, but with cleavage at slightly different peptide bonds. As shown in Figure 1B, trypsin initially cleaves the polymerase into 80K and 20K polypeptide fragments; however, with time the 20K fragment is slowly degraded. In contrast, the 80K fragment remains mostly uncleaved over the entire time course (Figure 1B). Gas-phase peptide sequencing of the product mixture reveals that after 10 min (when the protein is mostly 80K-20K) there are two predominant proteolysis sites, with cleavage after Arg-173 and after Lys-180.
V O L . 2 7 , NO. 1 5 , 1 9 8 8
5765
After extensive digestion with trypsin (90 min at room temperature), sequence and gel analyses of the product reveal complete cleavage occurring after Lys- 180, with extensive cleavage occurring within the 20K polypeptide. Although the region of the polypeptide between amino acid residue 172 and residue 179 is readily cleaved by proteases, we have been unable to separate the 20K and 80K fragments by any means short of denaturation. The two domains appear to be held together by strong hydrophobic interactions. In the following, we will refer to uncleaved enzyme as the native form and to the two proteolytically modified forms of T7 RNA polymerase as the 80K-20K form, containing a unique cleavage site following Lys-179 (produced by cleavage with whole cells), and the 80K form, in which the N-terminal 20K fragment (the region of the sequence preceding Ala-181) is degraded by extensive proteolysis with trypsin. Standard Activity Assay. Polymerase activity on whole T7 DNA was measured for the three different enzyme samples (native, 80K-20K, and 80K) as previously described (Chamberlin & Ring, 1973; Oakley et al., 1975). Native enzyme typically shows a standard activity of 300000 units/mg (1 unit = 1 nmol of ATP incorporated per hour at 37 "C). In this assay, the singly nicked species has been shown to possess approximately 30% the activity of native enzyme (Ikeda 8c Richardson, 1987a). Our results confirm this finding for the 80K-20K species, with a standard activity on whole T7 DNA of approximately 100000 units/mg. In contrast, the 80K species is completely inactive as judged by this assay (
